群组决策中残缺判断矩阵的相容性修正及排序新算法

判断矩阵的相容性检验及修正是AHP模型应用于群组决策排序的关键步骤.但目前许多文献仅提供了相容性指标的检验方法,而未考虑其修正算法.一些文献中虽然提出了判断矩阵相容性的修正算法,但对残缺矩阵、专家权重不同以及修正后矩阵的可信度等情况考虑不足,同时修正迭代步长的选取具有较大盲目性.针对这一现状,提出了基于可能满意度的残缺矩阵相容性修正及排序新算法.此算法很好地解决了现有算法中存在的问题和不足.最后通过算例证明算法的有效性.

A New Algorithm for Incomplete Matrixes' Compatibility Improvement and Group Ranking in Group Decision Making

Compatibility Improvement is a key step in AHP applied in group decision making. But many literatures only discussed how to verify the Compatibility.Although some papers provided methods to improve compatibility,they did not take enough factors into account,such as incomplete matrix,different weights between experts and the reliability of matrixes after amending.Owing to the shortcoming of existing methods,we put forward a new algorithm based on possibility-satisfiability degree.By this method,we could resolve the problems mentioned above very well. The final example proved the flow and efficient of the new method.